1.👨‍🔬Proud science nerd who in his spare time thinks about how geology🪨, biology🌳, and water💧interact
2. 🌐I build spatial systems that turn imagery, maps, and large datasets into info about the physical world
3. 👨‍💻I bring a scientist's mindset to engineering end-to-end software, from raw spatial data to ML models

I’m a spatial systems engineer and geologist by training. I like working across the full spatial data lifecycle: acquiring real-world data, structuring it for efficient analysis, building spatial logic, training models when ML is useful, and turning the result into reliable software. My work sits at the intersection of spatial data systems, remote sensing, software engineering, and machine learning.

Here are some of the things I've done:

• Architected an end-to-end spatial ML system using PostGIS, computer vision, and production inference to model building demolition risk
• Built satellite data systems using Sentinel-2, Xarray/Zarr, AWS S3, and PyTorch for spatiotemporal change detection
• Improved a production spatial-processing workflow by 40× through profiling and redesigning multiprocessing and logging bottlenecks
• Led production data-pipeline development and technical handoff while mentoring an intern on an applied spatial ML research project
• Presented my building-demolition research at the 2026 Geospatial World Forum in Amsterdam

I’m particularly interested in systems that use remote sensing, computer vision, and large-scale spatial computation to understand how the built and natural environment changes over time.

               

• PyMAAP
  • Code benchmarking and analysis tool available on PyPI
  • Package offers timing decorators and inline performance monitoring hooks
  • Script run from the command line can automatically generate helpful plots and tables so you can analyze the performance of your script(s)
• Sustainability Insights
  • Python-based project centered on gleaning insights from mining emissions datasets sourced from Climate TRACE
    • Python tools used: Numpy, Pandas, DuckDB, and Matplotlib
  • A Tableau Public Story was created to show which companies are most responsible for the world's mining emissions
  • I wrote an Executive Summary which explains my workflow and conclusions
• ClassiFIRE
  • Python- and QGIS-based wildfire prediction model
    • Python tools used: Numpy, Pandas, Matplotlib, Seaborn, Scikit-Learn, Tensorflow
  • Presentation of my findings
  • Executive Summary of my work
• Potential Talents
  • Python-based NLP project designed to utilize Learning-To-Rank systems to organize job candidates based on their similarity to particular terms
    • Visualization techniques used: histograms, boxplots, bar charts, choropleth map, word cloud
    • Python tools used: Text embedding (Tfidf, Word2Vec, GloVe, fastText, SBERT), Scikit-Learn's Cosine Similarity, RankNet with PyTorch, LambdaRank with LightGBM